Skyline preference query method based on mass incomplete data sets

Abstract

The invention relates to a skyline preference query method based on mass incomplete data sets. According to the method, the incomplete data sets IS are projected according to the importance of properties and preferences of users, stringent clustering and loose clustering are performed on two data sets IS' and IS" obtained through projection, two different skyline preference query algorithms are executed respectively after clustering to obtain a skyline result set SSRS based on stringent clustering and a skyline result set RSRS based on loose clustering, and finally a skyline preference query result selection strategy based on information entropy calculation is executed once to obtain a skyline query result set meeting the preferences of the users. Through the method, extraction of personalized information from the mass incomplete data sets is effectively realized, and the efficiency of the skyline query algorithms on the mass incomplete data sets is improved.

Description

technical field [0001] The invention relates to a skyline preference query method based on massive incomplete data sets, and belongs to the technical field of Internet of Things and big data processing. Background technique [0002] The Internet of things (IoT) is an important part of the new generation of information technology, and it is also an important development stage of informatization. At present, sensors and monitoring equipment are mainly used to obtain data in the field of the Internet of Things. Due to various situations such as sensor and monitoring equipment failures, errors, and actual data acquisition limitations, data understanding errors or data omissions, etc., the data set is not accurate. Completeness is ubiquitous. Such datasets with missing data are called incomplete datasets. With the development and popularization of Internet of Things applications, personalized recommendation aimed at meeting user needs has become a hot spot in Internet of Things...

AI Technical Summary

Problems solved by technology

However, preprocessing consumes too many system resources, and there are certain errors in the repaired data, resulting in inaccurate query results

And for some timeliness issues, such as data during the flu period, preprocessing these strong timeliness data may lead to data invalidation

Images